Model comparison
Claude Sonnet 4 vs Gemma 4 31B IT
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 40.8 on the Noometry Index.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Claude Sonnet 4 scores higher in 3 categories and Gemma 4 31B IT in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Gemma 4 31B IT leads 41.6 to 26.2.
- The biggest single-benchmark swing is LMCA: 29% for Claude Sonnet 4 and 39.3% for Gemma 4 31B IT.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Gemma 4 31B IT accepts more context: 262K tokens versus 200K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Gemma 4 31B IT | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 40.8 | 43.5 |
| Released | 2025-05-22 | 2026-04-02 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 64K | 33K |
| Input $ / M tokens | $3 | $0.09 |
| Output $ / M tokens | $15 | $0.34 |
| Results tracked | 58 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 43.5 (#88), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| SciCode | 40% | 43.4% |
| WeirdML | 46.1% | 52.3% |
| LMArena Coding | 1414 | 1459 |
| ALE-Bench | 655.35 | 925.5 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1366 |
| GSO | 4.9% | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), Gemma 4 31B IT: —
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Gemma 4 31B IT leads
Claude Sonnet 4: 22.9 (#187), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| Kagi LLM Benchmark | 73% | 63.5% |
| CritPt | 0.3% | 1.4% |
| LMArena Hard Prompts | 1372 | 1448 |
| DTBench | 77.1% | 82.7% |
| LMCA | 29% | 39.3% |
| Epoch Capabilities Index | 141.69 | 142.74 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| NYT Connections (extended) | — | 70.6% |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 5% |
| EnigmaEval | 3.1% | — |
| Thematic Generalization | — | 53% |
| Surface Evolver Bench | — | 30.6% |
| ForecastBench | 60.2 | — |
Math Too close to call
Claude Sonnet 4: 43.3 (#80), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 73.3% |
| LMArena Math | 1375 | 1465 |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.8 (#108), Gemma 4 31B IT: 37.9 (#151)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| GPQA Diamond | 79.2% | 75.8% |
| Vectara Hallucination Rate | 10.3% | 7.4% |
| LMArena Expert | 1372 | 1465 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 10.4% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal Gemma 4 31B IT leads
Claude Sonnet 4: 26.2 (#121), Gemma 4 31B IT: 41.6 (#34)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | 1191 | 1277 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| LMArena Document | — | 1425 |
| MindCube | 44.8% | — |
Multilingual Gemma 4 31B IT leads
Claude Sonnet 4: 46.7 (#156), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1333 | 1431 |
| LMArena Chinese | 1350 | 1476 |
| LMArena French | 1363 | 1435 |
| LMArena Russian | 1355 | 1460 |
| LMArena Spanish | 1357 | 1444 |
| LMArena German | 1331 | — |
| LMArena Japanese | 1302 | — |
| LMArena Korean | 1291 | — |
Instruction Following Gemma 4 31B IT leads
Claude Sonnet 4: 71.7 (#145), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1376 | 1433 |
| IFEval | 84% | — |
Long Context Gemma 4 31B IT leads
Claude Sonnet 4: 33.7 (#259), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1398 | 1446 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Gemma 4 31B IT leads
Claude Sonnet 4: 57.1 (#132), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | Claude Sonnet 4 | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1351 | 1443 |
| LMArena Creative Writing | 1345 | 1415 |
| EQ-Bench Creative Writing | 1483 | 1368 |
| LMArena Multi-Turn | 1376 | 1452 |
| Short-Story Creative Writing | 81.4% | — |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1120 |
Frequently asked questions
Is Claude Sonnet 4 better than Gemma 4 31B IT?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 40.8 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or Gemma 4 31B IT?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Gemma 4 31B IT better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 42.3 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Gemma 4 31B IT share?
27 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Gemma 4 31B IT has 35.